Method and apparatus for programming a natural neuronal network into solid-state electronic memories network and circuits
The device defect prevention enhancement system addresses cosmic ray-induced soft errors in supercomputers by dynamically adjusting checkpoint cycles and supply voltage based on cosmic ray event data, ensuring reliable operation and completion of computations.
Patent Information
- Application Number
- KR1020210116914
- Authority / Receiving Office
- KR · KR
- Patent Type
- Patents
- Current Assignee / Owner
- Filing Date
- 2021-09-02
- Publication Date
- 2026-11-11
- Estimated Expiration
- 2041-09-02
AI Technical Summary
Existing devices, particularly supercomputers, are prone to soft errors caused by cosmic rays, leading to increased failure rates and incomplete computations due to the growing number of nodes, which conventional checkpoint cycles and voltage management methods fail to adequately address.
A device defect prevention enhancement system that includes a processor to acquire cosmic ray event data, determine a soft error overcoming policy by considering MTBF, checkpoint cycles, supply voltage, and operating mode, and transmit this policy to the device, using an artificial neural network to predict event information based on location.
Enhances device reliability by dynamically adjusting checkpoint cycles and supply voltage, reducing soft error occurrences and ensuring continuous operation by anticipating and mitigating cosmic ray-induced failures.
Smart Images

Figure 112021101758470-PAT00003_ABST
Abstract
Description
Technology Field
[0001] The following embodiments relate to a method and apparatus for enhancing device defect prevention. Background Technology
[0003] Components that make up a device, such as the CPU, DRAM, and storage, may experience problems due to errors in the hardware itself, but they may also experience problems due to soft errors caused by cosmic rays coming from space. means of solving the problem
[0005] A device defect prevention enhancement device according to one embodiment includes the steps of: acquiring information regarding at least one event that may occur due to a cosmic ray; determining a soft error overcoming policy based on the information regarding the event; and transmitting the soft error overcoming policy to a device.
[0006] The step of determining the above soft error recovery policy may include: a step of determining the Mean Time Between Failure (MTBF) of the device by considering information regarding the above event; and a step of determining the checkpoint cycle of the device based on the mean time between failure.
[0007] The step of determining the average failure-free time may include receiving a first average failure-free time of the device; and modifying the first average failure-free time by taking into account information regarding the event.
[0008] The step of determining the above soft error overcoming policy may include the step of determining the magnitude of the supply voltage applied to the device by considering information regarding the above event.
[0009] The step of determining the above soft error overcoming policy may include a step of determining the operating mode of the device by considering information regarding the above event.
[0010] The step of obtaining information regarding the above event may include the step of obtaining data related to the spacecraft; and the step of inputting the data related to the spacecraft into an artificial neural network to predict information regarding the event that may occur.
[0011] The step of predicting information regarding the above event may include the step of correcting the output of the artificial neural network based on the location information of the device.
[0012] The step of predicting information regarding the above event may include: a step of predicting energy level information of the event according to the location of the device; a step of predicting period information of the event according to the location of the device; and a step of predicting arrival time information of the event according to the location of the device.
[0013] The step of obtaining information regarding the above event may include the step of receiving information regarding the above event.
[0014] A device defect prevention enhancement device according to one embodiment includes a processor that acquires information regarding at least one event that may occur due to a cosmic ray, determines a soft error overcoming policy based on the information regarding the event, and transmits the soft error overcoming policy to the device.
[0015] The processor can determine the Mean Time Between Failure (MTBF) of the device by considering information regarding the event, and determine the checkpoint period of the device based on the Mean Time Between Failure.
[0016] The processor receives a first average time without failure of the device and can modify the first average time without failure by taking into account information regarding the event.
[0017] The processor can determine the magnitude of the supply voltage applied to the device by taking into account information regarding the event.
[0018] The above processor can determine the operating mode of the device by considering information regarding the above event.
[0019] The processor can acquire data related to the spacecraft and input the data related to the spacecraft into an artificial neural network to predict information regarding the event that may occur.
[0020] The processor can correct the output of the artificial neural network based on the location information of the device.
[0021] The processor can predict energy level information of the event based on the location of the device, predict period information of the event based on the location of the device, and predict arrival time information of the event based on the location of the device.
[0022] The above processor can receive information regarding the above event. Brief explanation of the drawing
[0024] FIG. 1 is a diagram illustrating an example of an event that may occur due to a cosmic ray according to one embodiment. FIG. 2 is a flowchart illustrating a method for enhancing device defect prevention according to one embodiment. FIG. 3 is a diagram illustrating a method for obtaining information regarding an event according to the location of a device according to one embodiment. FIGS. 4a and FIGS. 4b are drawings for explaining the operation of a device defect prevention enhancement system according to one embodiment. FIG. 5 is a block diagram of a device defect prevention enhancement device according to one embodiment. Specific details for implementing the invention
[0025] The specific structural or functional descriptions disclosed in this specification are illustrative of embodiments according to technical concepts only, and the actual implemented form may take various other forms and is not limited to the embodiments described in this specification.
[0026] Terms such as "first" or "second" may be used to describe various components, but these terms should be understood solely for the purpose of distinguishing one component from another. For example, the first component may be named the second component, and similarly, the second component may be named the first component.
[0027] When it is stated that one component is "connected" or "connected" to another component, it should be understood that while it may be directly connected or connected to that other component, there may also be other components in between. Conversely, when it is stated that one component is "directly connected" or "directly connected" to another component, it should be understood that there are no other components in between. Expressions describing the relationships between components, such as "between" and "directly between," or "adjacent to" and "directly adjacent to," should be interpreted in the same way.
[0028] The singular expression includes the plural expression unless the context clearly indicates otherwise. In this specification, terms such as “comprising” or “having” are intended to specify the existence of the implemented features, numbers, steps, actions, components, parts, or combinations thereof, and should be understood as not precluding the existence or addition of one or more other features, numbers, steps, actions, components, parts, or combinations thereof.
[0029] Unless otherwise defined, all terms used herein, including technical or scientific terms, have the same meaning as generally understood by those skilled in the art. Terms such as those defined in commonly used dictionaries should be interpreted as having a meaning consistent with their meaning in the context of the relevant technology, and should not be interpreted in an ideal or overly formal sense unless explicitly defined in this specification.
[0030] The embodiments can be implemented in various forms of products, such as personal computers, laptop computers, tablet computers, smartphones, televisions, smart home appliances, intelligent vehicles, kiosks, and wearable devices. The embodiments will be described in detail below with reference to the attached drawings. Identical reference numerals in each drawing indicate identical components.
[0032] FIG. 1 is a diagram illustrating an example of an event that may occur due to a cosmic ray according to one embodiment.
[0033] According to one embodiment, cosmic rays are a general term for high-energy particles having various and large amounts of energy. As the protons of the cosmic rays descend to the ground, they decay into pion mesons, and furthermore, the mesons can undergo a decay process into electrons (or positrons), neutrinos, photons, etc., through muon particles which have a lighter mass.
[0034] A cosmic ray according to one embodiment may include a galactic cosmic ray and a solar cosmic ray. For convenience of explanation, the following description is based on a solar cosmic ray.
[0035] Referring to Figure 1, various solar activities (e.g., flares, solar proton events, coronal mass ejections, high-speed solar winds) can generate cosmic rays, and if the device is exposed to cosmic rays, various events that cause soft errors may occur. For example, if a powerful sunspot explosion occurs, light such as X-rays and ultraviolet rays first reaches Earth in about 8 minutes, which may cause events such as the device's shortwave communication interruption and satellite reception performance degradation.
[0036] According to one embodiment, a solar spacecraft can be significantly altered by solar activity. For example, coronal mass ejections occur due to sunspots, and the size of the coronal mass ejection during the maximum sunspot period is 50 times greater than during the minimum sunspot period. When solar activity is greatly intensified in this way, the occurrence of problems caused by soft errors may increase.
[0037] In particular, supercomputers are continuously growing in size to solve increasingly complex problems (Single Large Problems). In other words, the number of nodes is continuously increasing. The number of nodes, which was previously only a few hundred, has grown to over 100,000, and the number of nodes required to solve complex problems is also increasing. However, as the number of nodes increases, the occurrence of problems caused by soft errors also increases, and if a soft error occurs unexpectedly during computation, the program may not be able to be completed.
[0038] Below, with reference to FIGS. 2 to 5, a method to overcome soft errors that may occur due to a spacecraft is described in detail.
[0040] FIG. 2 is a flowchart illustrating a method for enhancing device defect prevention according to one embodiment.
[0041] Referring to FIG. 2, steps (210 to 230) may be performed by a device defect prevention enhancement device according to one embodiment. A device defect prevention enhancement device according to one embodiment may be implemented by one or more hardware modules, one or more software modules, or various combinations thereof.
[0042] The operations of FIG. 2 may be performed in the order and manner depicted, but the order of some operations may be changed or some operations may be omitted without departing from the spirit and scope of the depicted embodiment. A number of operations depicted in FIG. 2 may be performed in parallel or simultaneously.
[0043] In step (210), a device fault prevention enhancement device according to one embodiment obtains information regarding at least one event that may occur due to a spacecraft. As will be described in detail below, the device fault prevention enhancement device according to one embodiment may receive data related to the spacecraft and directly predict information regarding the event, or may receive information regarding the event from another entity (e.g., a space activity analyzer).
[0044] Information regarding an event according to one embodiment may include information on the energy level of the event, information on the period of occurrence of the event, and / or information on the time to arrive of the event. Furthermore, information regarding an event according to one embodiment may vary depending on the location of the device. The location of the device may be expressed in latitude and longitude.
[0045] A device defect prevention enhancement device according to one embodiment can obtain information regarding events in the form of a table. Table 1 below is an example of information regarding events in the form of a table.
[0046]
[0047] In step (220), the device fault prevention enhancement device according to one embodiment determines a soft error overcoming policy based on information regarding an event. The device fault prevention enhancement device according to one embodiment determines the Mean Time Between Failure (MTBF) of the device by considering information regarding the event, and can determine the checkpoint cycle of the device based on the Mean Time Between Failure.
[0048] More specifically, checkpointing technology can be a technique that continuously saves the device's context periodically to recover and utilize that context when unexpected system errors occur.
[0049] That is, the device fault prevention enhancement device according to one embodiment saves the intermediate state of a program by performing checkpoints at regular time intervals in preparation for the occurrence of such errors, and enables the program to restart from the middle even if an error occurs. At this time, it is important to determine a reasonable checkpoint cycle considering system overhead and availability. The device fault prevention enhancement device according to one embodiment can dynamically determine the checkpoint cycle by reflecting soft errors that may occur due to spacecraft, which are not included in existing checkpoint cycle determination methods.
[0050] According to one embodiment, the checkpoint period may be determined based on the average failure-free time of the device. The checkpoint period must be shorter than the average failure-free time of the device to increase the efficiency of recovery following an error. The average failure-free time represents the average time for failures to occur in the device, and the average failure-free time in a given period may be calculated by dividing the total operating time during that period by the total number of failures. For example, when 3 failures occur during 600 hours, the average failure-free time for that period may be 200 (600 / 3).
[0051] Conventional average time-to-failure (MTF) is calculated by approximating the failure rate for soft errors to zero and based on hardware errors. Therefore, a checkpoint cycle determined based on the conventional MTF may not be suitable in situations where soft errors occur frequently. Accordingly, a device defect prevention enhancement device according to one embodiment can determine an MTF that reflects soft errors and determine a soft error overcoming policy including a checkpoint cycle based thereon.
[0052] A device defect prevention enhancement device according to one embodiment receives a first average failure-free time of a device and can modify the first average failure-free time by considering information regarding an event. The first average failure-free time may be an average failure-free time calculated based on hardware errors without considering soft errors. The device defect prevention enhancement device can determine a final average failure-free time that reflects soft errors by modifying the first average failure-free time.
[0053] More specifically, a device failure prevention enhancement device according to one embodiment can obtain an average failure-free time table of a device according to the flux of cosmic rays. The average failure-free time table of a device according to the flux of cosmic rays can be prepared in advance through an acceleration experiment. An acceleration experiment is an experiment in which a device is artificially irradiated with a large amount of neutrons and protons to measure how often it will fail due to cosmic rays; the device failure prevention enhancement device may generate the average failure-free time table directly through an acceleration experiment, or it may receive an average failure-free time table generated by another device.
[0054] A device defect prevention enhancement device according to one embodiment that has obtained an average time-by-time table can obtain an average time-by-time corresponding to information regarding an event (e.g., flux information of a spacecraft received at the current device location) by using the average time-by-time table.
[0055] A device fault prevention enhancement device according to one embodiment can determine a checkpoint period based on average time without failure. If soft errors caused by neutrons increase significantly, the checkpoint period can be shortened (e.g., existing checkpoint period: once every 300 steps, changed checkpoint period: once every 180 steps).
[0056] Furthermore, a device defect prevention enhancement device according to one embodiment can determine the magnitude of the supply voltage applied to the device by considering information regarding an event. A device defect prevention enhancement device according to one embodiment can increase the voltage applied to the device within an allowable voltage so that the device does not discharge when a soft error increases significantly.
[0057] Furthermore, a device defect prevention enhancement device according to one embodiment can determine the operating mode of the device by considering information regarding an event. For example, if a soft error increases significantly, the device defect prevention enhancement device according to one embodiment can change the operating mode of the device from a performance mode to a reliability mode.
[0058] In step (230), the device defect prevention enhancement device according to one embodiment transmits a soft error overcoming policy to the device. The device according to one embodiment may include various types of products such as a supercomputer, personal computer, laptop computer, tablet computer, smartphone, television, smart home appliance, intelligent vehicle, kiosk, and wearable device. However, for convenience of explanation below, the description is based on a supercomputer.
[0059] A supercomputer in the form of a large-scale parallel system cluster including a plurality of nodes according to one embodiment may include a job scheduler and may receive a soft error tolerance policy through the job scheduler. The supercomputer that receives the soft error tolerance policy may transmit the soft error tolerance policy to each node. The supercomputer according to one embodiment determines whether each node is currently operating, and if it is operating, it may normally stop the currently running job and change the checkpoint cycle; if it is not operating, it may change the checkpoint cycle to be found when the job proceeds later. Furthermore, the supercomputer according to one embodiment may change voltage conditions, operating modes, etc., according to the soft error tolerance policy as well as the checkpoint cycle. For example, the supercomputer according to one embodiment may increase the voltage applied to components within the entire server.
[0060] A device defect prevention enhancement device according to one embodiment can determine whether to transmit a soft error overcoming policy related to an event by comparing the arrival time of an event with the sum of the time taken to acquire data related to the spacecraft and the time taken to predict information regarding the event.
[0061] For example, if it takes 1 day for the solar wind to reach the device, 8 minutes for image observation, and 15 minutes for image analysis, the policy to overcome soft errors caused by the solar wind can be delivered to the device because image observation and analysis are possible before the solar wind reaches the device.
[0062] On the other hand, if X-rays and ultraviolet rays reach the device in 8 minutes due to a powerful sunspot explosion, the soft error overcoming policy associated with the event may not be delivered to the device because the X-rays and ultraviolet rays reach the device before image observation and analysis are completed.
[0064] FIG. 3 is a diagram illustrating a method for obtaining information regarding an event according to the location of a device according to one embodiment.
[0065] Referring to FIG. 3, an artificial neural network (320) according to one embodiment can receive data (310) related to a spacecraft and output output data (330).
[0066] An artificial neural network (320) according to one embodiment may include an input layer, a hidden layer, and an output layer. Each layer may include a plurality of nodes, and nodes between adjacent layers may be connected to each other with connection weights. Each node may operate based on an activation model. An output value corresponding to an input value may be determined according to the activation model. The output value of any node may be input to a node of the next layer connected to that node. A node of the next layer may receive values output from a plurality of nodes. During the process of inputting the output value of any node to a node of the next layer, a connection weight may be applied. A node of the next layer may output an output value corresponding to an input value based on the activation model to a node of the next layer connected to that node. An output layer may include nodes corresponding to a plurality of elements. Nodes of the output layer may output feature values corresponding to a plurality of elements.
[0067] Model parameters refer to parameters determined through learning and may include synaptic connection weights and neuron biases. Hyperparameters, on the other hand, refer to parameters that must be set prior to training in a machine learning algorithm and may include the learning rate, number of iterations, mini-batch size, and initialization function.
[0068] The purpose of training an artificial neural network (320) can be seen as determining model parameters that minimize the loss function. The loss function can be used as an indicator to determine optimal model parameters during the training process of the artificial neural network.
[0069] An artificial neural network (320) according to one embodiment can be trained based on a pair of training data related to a spacecraft (e.g., a real-time collected corona graph) and corresponding event data (e.g., coronal mass ejection amount). More specifically, the artificial neural network (320) can determine the difference between the predicted output data and the correct data based on the training data related to the spacecraft as a loss function, and can determine model parameters that minimize the loss function.
[0070] The arrival time and effect of an event may vary depending on the location of the device according to one embodiment. However, the output data (330) according to one embodiment may be data that does not take the location of the device into account.
[0071] A device defect prevention enhancement device (or space activity analyzer) according to one embodiment can obtain the device's location information data (340) by correcting the output data (330) of an artificial neural network (320) based on the device's location information data (340).
[0072] The location information data (340) of a device according to one embodiment may include energy level information according to the location of the device, period information of an event, and / or arrival time information of an event.
[0074] FIGS. 4a and FIGS. 4b are drawings for explaining the operation of a device defect prevention enhancement system according to one embodiment.
[0075] A device fault prevention enhancement device according to one embodiment may receive data related to a spacecraft and predict information regarding an event directly, or receive information regarding an event from another entity (e.g., a space activity analyzer). Below, FIG. 4a illustrates an embodiment in which information regarding an event is received from a space activity analyzer, and FIG. 4b illustrates an embodiment in which the device fault prevention enhancement device directly predicts information regarding an event. Since the description of FIG. 1 to 3 is applicable to FIG. 4a to 4b, redundant content may be omitted.
[0076] Referring to FIG. 4a, a device defect prevention enhancement system according to one embodiment may include an observatory (411), a space activity analyzer (412), a server (413), and a device (414).
[0077] An observatory (411) according to one embodiment can obtain data related to spacecraft by observing celestial activity including the sun. For example, the observatory (411) can obtain the number of sunspots, corona graphs, etc. using a camera, a satellite, etc.
[0078] A space activity analyzer (412) according to one embodiment can predict information regarding events that may occur by inputting data related to spacecraft into an artificial neural network. For example, a space activity analyzer (412) according to one embodiment can predict the amount of spacecraft entering from space based on data related to spacecraft.
[0079] A server (413) according to one embodiment may determine a soft error overcoming policy based on information regarding an event and transmit the soft error overcoming policy to a device (414). A server (413) according to one embodiment may be a device defect prevention enhancement device described above with reference to FIGS. 1 to 3.
[0080] Referring to FIG. 4b, a device defect prevention enhancement system according to another embodiment may include an observation station (421), a server (422), and a device (423) as the main components.
[0081] The observatory (421) and device (423) according to one embodiment can each operate in the same way as the observatory (411) and device (414) of FIG. 4a. The server (422) according to one embodiment can receive data related to the spacecraft from the observatory (421) and directly predict information regarding the event.
[0083] FIG. 5 is a block diagram of a device defect prevention enhancement device according to one embodiment.
[0084] Since the device defect prevention enhancement device according to one embodiment can secure a registered voice during a speaker recognition operation, it may be referred to as a device defect prevention enhancement device (500) for speaker recognition.
[0085] Referring to FIG. 5, a device defect prevention enhancement device (500) according to one embodiment includes a processor (510). The device defect prevention enhancement device (500) may further include a memory (530), a communication interface (550), and sensors (570). The processor (510), memory (530), communication interface (550), and sensors (570) may communicate with each other through a communication bus (505).
[0086] The processor (510) obtains information regarding at least one event that may occur due to a cosmic ray, determines a soft error overcoming policy based on the information regarding the event, and transmits the soft error overcoming policy to the device.
[0087] The memory (530) can store information regarding at least one event that may occur due to a cosmic ray, and the average failure-free time table of the device. The memory (530) may be volatile memory or non-volatile memory.
[0088] The sensors (570) may include, for example, camera sensors for acquiring data related to the spacecraft.
[0089] In addition, the processor (510) may perform at least one method or an algorithm corresponding to at least one method described above through FIGS. 1 to 4b. The processor (510) may execute a program and control the device fault prevention enhancement device (500). The program code executed by the processor (510) may be stored in memory (530). The device fault prevention enhancement device (500) may be connected to an external device (e.g., a personal computer or a network) through an input / output device (not shown) and may exchange data. The device fault prevention enhancement device (500) may be installed on various computing devices and / or systems, such as smartphones, tablet computers, laptop computers, desktop computers, televisions, wearable devices, security systems, smart home systems, etc.
[0091] The embodiments described above may be implemented as hardware components, software components, and / or combinations of hardware and software components. For example, the devices, methods, and components described in the embodiments may be implemented using a general-purpose computer or a special-purpose computer, such as, for example, a processor, a controller, an arithmetic logic unit (ALU), a digital signal processor, a microcomputer, a field programmable gate array (FPGA), a programmable logic unit (PLU), a microprocessor, or any other device capable of executing and responding to instructions. The processing unit may execute an operating system (OS) and software applications executed on said operating system. Additionally, the processing unit may access, store, manipulate, process, and generate data in response to the execution of the software. For ease of understanding, the processing unit may be described as being used as a single unit, but those skilled in the art will understand that the processing unit may include multiple processing elements and / or multiple types of processing elements. For example, the processing unit may include multiple processors or one processor and one controller. Additionally, other processing configurations, such as parallel processors, are also possible.
[0092] Software may include computer programs, code, instructions, or a combination thereof, and may configure a processing unit to operate as desired or command the processing unit independently or collectively. Software and / or data may be permanently or temporarily embodied in any type of machine, component, physical device, virtual equipment, computer storage medium or device, or transmitted signal wave in order to be interpreted by the processing unit or to provide instructions or data to the processing unit. Software may be distributed over networked computer systems and may be stored or executed in a distributed manner. Software and data may be stored on computer-readable recording media.
[0093] The method according to the embodiment may be implemented in the form of program instructions that can be executed through various computer means and recorded on a computer-readable medium. The computer-readable medium may include program instructions, data files, data structures, etc., either alone or in combination. The program instructions recorded on the medium may be those specifically designed and configured for the embodiment, or they may be those known and available to those skilled in the art of computer software. Examples of computer-readable recording media include magnetic media such as hard disks, floppy disks, and magnetic tapes; optical recording media such as CD-ROMs and DVDs; magneto-optical media such as floptical disks; and hardware devices specifically configured to store and execute program instructions, such as ROM, RAM, and flash memory. Examples of program instructions include machine code, such as that generated by a compiler, as well as high-level language code that can be executed by a computer using an interpreter, etc.
[0094] Although the embodiments have been described above with reference to the limited drawings, those skilled in the art can apply various technical modifications and variations based on the above. For example, suitable results may be achieved even if the described techniques are performed in a different order than described, and / or if the components of the described system, structure, device, circuit, etc. are combined or assembled in a form different from described, or replaced or substituted by other components or equivalents.
[0095] Therefore, other implementations, other embodiments, and equivalents to the claims also fall within the scope of the claims set forth below.
Claims
Claim 1 A method for enhancing device defect prevention performed by a processor, comprising: a step of acquiring information regarding at least one event that may occur due to a cosmic ray; a step of determining a soft error overcoming policy based on the information regarding the event; and a step of transmitting the soft error overcoming policy to a device, wherein the step of determining the soft error overcoming policy includes a step of determining the Mean Time Between Failure (MTBF) of the device by considering the information regarding the event; and a step of determining a checkpoint period of the device based on the mean time between failure; wherein the step of determining the mean time between failure includes a step of receiving a first mean time between failure of the device; and a step of modifying the first mean time between failure by considering the information regarding the event. Claim 2 delete Claim 3 delete Claim 4 A method for enhancing device defect prevention according to claim 1, wherein the step of determining the soft error overcoming policy includes the step of determining the magnitude of the supply voltage applied to the device by considering information regarding the event. Claim 5 A method for enhancing device defect prevention according to claim 1, wherein the step of determining the soft error overcoming policy includes the step of determining the operating mode of the device by considering information regarding the event. Claim 6 A method for enhancing device defect prevention according to claim 1, wherein the step of obtaining information regarding the event comprises: a step of obtaining data related to the spacecraft; and a step of inputting the data related to the spacecraft into an artificial neural network to predict information regarding the event that may occur. Claim 7 In claim 6, the step of predicting information regarding the above event includes the step of correcting the output of the artificial neural network based on the location information of the device, a method for enhancing device defect prevention. Claim 8 In claim 6, the step of predicting information regarding the event comprises: a step of predicting energy level information of the event according to the location of the device; a step of predicting period information of the event according to the location of the device; and a step of predicting arrival time information of the event according to the location of the device, a method for enhancing device defect prevention. Claim 9 A method for enhancing device defect prevention, wherein, in claim 1, the step of obtaining information regarding the event includes the step of receiving information regarding the event. Claim 10 A computer program stored on a medium in combination with hardware to execute the method of any one of claims 1 and 4 through 9. Claim 11 A device fault prevention enhancement device comprising a processor that acquires information regarding at least one event that may occur due to a cosmic ray, determines a soft error overcoming policy based on the information regarding the event, and transmits the soft error overcoming policy to a device, wherein the processor determines the Mean Time Between Failure (MTBF) of the device by considering the information regarding the event, determines a checkpoint period of the device based on the Mean Time Between Failure, receives a first Mean Time Between Failure of the device, and modifies the first Mean Time Between Failure by considering the information regarding the event. Claim 12 delete Claim 13 delete Claim 14 In claim 11, the device defect prevention enhancement device, wherein the processor determines the magnitude of the supply voltage applied to the device by taking into account information regarding the event. Claim 15 In claim 11, the above processor determines the operating mode of the device by considering information regarding the above event, a device defect prevention enhancement device. Claim 16 In claim 11, the device defect prevention enhancement device, wherein the processor acquires data related to the spacecraft and inputs the data related to the spacecraft into an artificial neural network to predict information regarding the event that may occur. Claim 17 In claim 16, the above processor corrects the output of the artificial neural network based on the location information of the device, a device defect prevention enhancement device. Claim 18 In claim 16, the device defect prevention enhancement device, wherein the processor predicts energy level information of the event according to the location of the device, predicts period information of the event according to the location of the device, and predicts arrival time information of the event according to the location of the device. Claim 19 In paragraph 11, the processor is a device fault prevention enhancement device that receives information regarding the event.
Citation Information
Patent Citations
Simulation method and system for predicting ser
KR1020180046105A
Detection mechanism
US20060010346A1
On-die electric cosmic ray detector
US20150355343A1
Proactive Failure Recovery Model for Distributed Computing
US20160034362A1
Apparatus and method for optimizing reliability of satellite system considering both hard error and soft error
KR102263947B1
